Challenging misleading healthcare claims.

The Society of Homeopaths: failing to make the case for homeopathy

Homeopaths publish ten studies they claim affirm the ‘potential benefits of homeopathy for a range of health problems’, but do they withstand scrutiny?

sugarpills1Homeopathy is under severe pressure in the UK with what little credibility it had, destroyed. The ending of referrals from GPs in Liverpool and the Wirral over the last year or so and consultations on the future of homeopathy in Clinical Commissioning Groups in Bristol and Enfield, and more recently by NHS England, have piled on top of the decline by 96% of NHS prescriptions for homeopathy in England in the past 20 years.

We have also seen the recent announcement that Weleda (a supplier of homeopathic and anthroposophic products to the NHS) is ending the production of their ‘bespoke’ homeopathic products. The pending outcome of a consultation by the Charity Commission on CAM charities could add yet more woe.

We could be seeing the final death throes of homeopathy on the NHS with possible knock-on effects on the businesses of lay homeopaths: many relying on the false imprimatur leant to it by the NHS.

In November, the Council of the Royal College of Veterinary Surgeons (RCVS) issued a position statement including ‘Homeopathy exists without a recognised body of evidence for its use. Furthermore, it is not based on sound scientific principles.’

This followed a statement in September 2017 from the European Association of Science Advisory Councils, which represents the national science academies of 27 countries, concluding that homeopathy ‘can actually be harmful: by delaying or deterring a patient from seeking appropriate, evidence-based, medical attention and by undermining patient and public confidence in scientific evidence.’

Homeopaths are also under pressure from the Advertising Standards Authority to make sure their advertising is ‘legal, decent, honest and truthful’, but many still seem to have difficulty with those principles. To protect the public from misleading claims, the ASA rightly sets a high standard of evidence — one clearly too high for homeopaths. But the ASA make it easy for homeopathy advertisers, stating: “Practitioners should therefore avoid making direct or implied claims that homeopathy can treat medical conditions.” Crystal clear.

Whether it concerns human or animal health, the debate is clearly about the consideration of the best available scientific evidence; to that battle, homeopaths come unarmed.


Homeopaths have brought much of this on themselves yet are keen to blame others. Failures to substantiate claims they make are met with a plethora of excuses. The trade bodies try to defend their members’ interests and businesses yet, for example, the response to the NHS England consultation from the Faculty of Homeopathy ran to a full four pages without citing a single scientific study. The Society of Homeopaths managed five pages and 18 references. The Homeopathy Research Institute managed eight pages and a handful of references but much of that attacked irrelevant issues and provided nothing new in the way of evidence to support homeopathy.

After the RCVS statement, homeopathy supporters have taken to petitions, postcards and even poetry in an attempt to persuade the RCVS that homeopathy does, indeed, work, despite the best scientific evidence to the contrary.

Horses have bolted and stable doors have been firmly closed and locked. We now only await the decision of Bristol CCGs who had been waiting for the NHS England decision. Meanwhile, the British Homeopathic Association say they are applying for a Judicial Review of, well, something or other they didn’t like about the NHS England consultation. And they’ve started a Parliamentary petition. All too little, all too late.

Research round-up

However, the Society of Homeopaths (SoH), keen that their members put forward the right message to anyone who might still be listening, launched a new campaign recently. Titled My Voice, it includes a new logo and a ‘research round-up’ they say will ‘help members make case for homeopathy’. They didn’t seem too keen to make this document public, but we have obtained a copy: Research studies for reference

In it, they make ten points about different aspects of homeopathy, all very positive, of course.

But do the claims they make in their headlines and in the details stand up to scrutiny? We analyse the ten papers below.

The common thread that runs through the papers is that their 'evidence' is, being charitable, weak and certainly not compelling. At worst, it is misunderstood, misinterpreted and misrepresented. It could be argued that they are simply incapable of understanding what differentiates good evidence from poor evidence. However, given that the SoH presents itself as an organisation working to the ‘highest professional standards’ in what they claim is ‘holistic medicine’, and one that boasts of its accreditation by the Professional Standards Authority, it really should be able to tell the difference if it is at all interested in protecting the public.

The papers

The ten supposedly affirm ‘the potential benefits of homeopathy for a range of health problems’.

As we see, the claims they make for these papers do no such thing. They misunderstand and mis-state the research they cite and do not give the complete picture of what these papers say.

Out of ten studies presented, presumably, as the most compelling they have, we have just one (number 6) that is a controlled trial, but its favourable evidence is weak and unreplicated.

Unfortunately, we see this all too often and we expect to see homeopaths cite the SoH document and their soundbites as gospel and as if they somehow counter robust scientific evidence. Readers now have the ammunition to counter those claims.


A fuller discussion of each paper and whether the claims made for them by the SoH are justified is given below, but first, a simple summary of the most obvious flaws of each paper:

1 Witt et al 2005: Poor methodology, patients also used conventional medicines, prone to bias and placebo effects.

2 Rossi et al 2014: Conflicts of interest (journal), no control, no placebo, no randomisation, high drop-out rate. The authors state the study 'does not allow to confirm the efficacy of homoeopathic therapies'.

3 Bornhöft et al 2011 (The Swiss Report): Conflict of interest (authors), not a Health Technology Assessment as claimed, simply a limited review of some homeopathy studies, the results of studies looked at were ‘re-interpreted' to create a more favourable impression of homeopathy, only one type of condition considered, the official report did not make the positive conclusions claimed by homeopathic proponents.

4 Grimaldi-Bensouda et al 2014: An observational study, not designed to say whether homeopathy is effective, potential vested interest (funding), no placebo group, conditions assessed were self-limiting and the use of homeopathic remedies is irrelevant to any reduction in the use of antibiotics.

5 Marian et al 2008: An unbalanced comparison of patients treated by homeopathic and conventional doctors, many other factors likely to account for the apparent better level of satisfaction with homeopaths, many 'homeopathic' patients were treated with conventional medicine, bias likely as homeopathic physicians and patients may have felt pressure to report favourable results, homeopathic patients described a lower level of symptom relief.

6 Del Carmen Macías-Cortés et al 2015: On the face of it a good study but small in size. Given this, the wider body of evidence and the unlikely nature of homeopathy, until this trial replicated it is likely to be an outlier with results attributable to chance.

7 To et al 2017: Unbalanced treatment of the two groups under comparison with the homeopathic group receiving more care and attention, no randomisation with potentially important group differences, reported outcomes unlikely to be related to homeopathic remedies, the SoH couldn't even get the authors' names right!

8 Viksveen et al 2017: Self-reported results, unblinded, no placebo control, so high risk of bias and placebo effect, small effect size, participants also received conventional medication, the A+B vs A design will always give an apparently positive result.

9 Frei 2015: No control group, so no allowance for the natural history of disease, observations likely to be biased and placebo effect, simply a series of case reports.

10 Frass et al 2015: No blinding, patients aware of the purpose of the study so expecting positive outcomes, homeopathic patients given extended consultation times compared with comparator group, subjective impressions only so prone to bias and placebo effect, the study was not designed to show which component of the homeopathic treatment made a difference.

The papers

Paper 1

SoH headline

Homeopathic medical treatment may play a beneficial role in the long-term care of patients with chronic diseases


A 2005 prospective, multicentre cohort study which reviewed the use of homeopathy found the severity of patients’ symptoms decreased significantly in the first two years of the trial. The most significant improvements were seen in the young and those with more severe symptoms. 3,981 patients consulting the doctor for the first time in 103 German and Swiss primary care practices were observed. Of these, 97% had chronic ailments lasting on average around 8 years. The most frequent ailments were allergic rhinitis in men, headache in women, and allergic skin conditions in children.


Witt CM, Lüdtke R, Baur R, Willich SN. Homeopathic medical practice: long-term results of a cohort study with 3981 patients. BMC Public Health [Internet]. 2005 Jul 1;5. Available at:


This self-reported study concluded that people with a long-standing chronic disease are the sort of people likely to seek out homeopathy. While the study was of a very good size and had a good response rate for a study on homeopathy, the methodology was poor. The patients were allowed to seek out conventional treatments while being part of this study. Also, the data collection was infrequent and allowing for the nature of chronic diseases and human memory this would introduce bias depending on how the questionnaire was ordered and the wording of the questions.

At best, they claim that homeopathy may play a beneficial role. This is not unusual for studies of homeopathy because efficacy can rarely be reliably established. Like most studies that show a positive outcome for Homeopathy this study has no control group which makes any claim of efficacy implausible. Chronic diseases, by their nature, are persistent, long-term, and susceptible to placebo effects.


Paper 2

SoH headline

Homeopathy can successfully treat a wide range of conditions


In a large-scale Italian study of over 2,592 patients between 1998 and 2010, 47% of patients reported major improvements or even resolution of their clinical picture, while only 10% saw no improvement. The most common presenting symptoms were respiratory (29.4%), digestive tract (17%) and skin complaints (14.7%).


Rossi E, Marco P, Paola B, Marialessandra P, Mariella DS, Monia P, et al. Homoeopathy in the public health system: Outcome data from the homoeopathic clinic of the Campo di Marte Hospital, Lucca, Italy (1998-2010). Eur J Integr Med. 2014;6(1):39–47. Available at:

[Note that the url provided by the SoH is wrong. The correct url is]


This study is published in the European Journal of Integrated Medicine so has a risk of bias due to vested interest. It is simply a large scale customer satisfaction survey compiled from responses from patients who knew they were receiving homeopathy and had elected to do so. Thus, from the start, the participants were primed to expect an improvement. This virtually guarantees the inclusion of every sort of cognitive bias (expectation, placebo effects and so on) that rigorous scientific trials are designed to correct for.

Many of the conditions that improved were self limiting and/or mild in nature, and many would be made worse by stress, so it is likely the lengthy (about half an hour) homeopathic consultations and regular reviews would have brought positive benefit in any case, without the need for homeopathy. The reader is unable to be certain however as there was no comparison made between homeopathic and conventional treatment. Even so, it took most of the patients who reported major improvement or resolution a full two years to get to that point.

Limitations of the trial as stated by the authors are that it had no control group (so obviously blinding was not considered) or randomisation. There was also a high attrition, or drop-out, rate that the authors (for reasons that were not given) assumed was because the homeopathy had worked and the patients felt no need to come back. In fact it is widely recognised high drop-out rates are often associated with patient dissatisfaction over a treatment not working.

The authors recognise these problems in a disclaimer towards the end, 'The high attrition rate, the absence of a control group, and randomization and complexity of the sample… does not allow to confirm the efficacy of homoeopathic therapies'.

Regarding the claim by the SoH that Homeopathy can successfully treat a wide range of conditions this is not supported in any way by this study and is directly contradicted by the authors who report, ‘The aim was, not to attribute any “cause and effect” to the intervention, but to investigate the reported outcomes of a sample of patients…’

In the final discussion, the authors state the results '…need to be confirmed with other controlled observational studies, on larger samples of patients, preferably using a randomized controlled trials design…', and I couldn't agree more. This paper is purely a propaganda exercise and the SoH claim regarding it is wrong.

Niall Taylor

Paper 3

SoH headline

Homeopathy is clinically effective, cost-effective and safe


This was the conclusion of a Health Technology Assessment report on homeopathy commissioned by the Swiss government. The report exhaustively reviewed the scientific literature and its unambiguous, positive conclusions have resulted in homeopathy being included on the list of medical treatments which are reimbursed through Switzerland’s national health insurance scheme.


Bornhöft G, Matthiessen PF. Homeopathy in Healthcare – Effectiveness, Appropriateness, Safety, Costs. An HTA report on homeopathy as part of the Swiss Complementary Medicine Evaluation Programme. Berlin Heidelberg: Springer-Verlag; 2011. Available at,%20Appr opriateness,%20Safety,%20Costs.pdf


The Swiss report on homeopathy did not show that homeopathy was ‘clinically effective, cost-effective and safe’.

It was not a Health Technology Assessment, which is a very specific, comprehensive and broad-ranging type of assessment, but simply a limited review of some homeopathy studies and with a slightly greater focus on homeopathy for a few medical conditions. Indeed, the Swiss Government felt it necessary to correct this false and misleading impression.

Further, the published document that the SoH cites wasn't even the same as that submitted to the Programm Evaluation Komplementärmedizin (PEK) set up by the Swiss Government to review reimbursement for homeopathy in their state health insurance scheme. It was an unofficial English translation of the report that was added to by the authors. Again, the Swiss Government had to clarify that it was published 'without any consent of the Swiss government or administration'.

In their evaluation of the evidence for homeopathy, the authors (mostly with homeopathy conflicts of interest) chose to 're-interpret' the conclusions of the original studies, making them more favourable to homeopathy. When the PEK reviewed the report, they had to downgrade the conclusions to achieve a more balanced view of the evidence, saying, 'Even less skeptical academic doctors will regard many interpretations as very optimistic and not scientifically convincing.'[1] It has been heavily criticised elsewhere, eg here and here, with the latter going as far as to call it ‘research misconduct’.

The report only looked specifically at evidence for upper respiratory tract infections and allergic reactions, so even if their conclusions had been valid, they cannot be extrapolated to the homeopathic treatment of any other condition. Also, homeopathy had not been found cost-effective because it had not been found effective.

As a direct result of the Swiss homeopathy report, the Government removed the previous temporary reimbursement of homeopathy from its insurance scheme.

However, after campaigning by homeopaths and their supporters, a referendum was held in which 67% of those voting voted for homeopathy (and other alternative treatments) to be included for reimbursement. The Swiss Government was in a difficult situation because, although the results of referendums are binding, their law only allows reimbursement for treatments that meet requirements of 'efficacy, appropriateness and cost-effectiveness'. Since none of these therapies met the requirements, the Government allowed temporary reimbursement until 2017. It has since been added back in again, but not because homeopathy met the requirements: the Swiss Government acknowledged that it was “impossible to provide such proof for these disciplines in their entirety”. It should also be noted that homeopathy can only be prescribed in Switzerland by a medical physician, not a lay homeopath such as might be a member of the SoH in the UK.

Alan Henness

Paper 4

SoH headline

The use of homeopathy could help to reduce the use of antibiotics and antipyretic/anti-inflammatory drugs for upper respiratory tract infections (URTIs).


Patients with URTIs given homeopathy in a four-year population-based study (EPI3 survey) of 825 GPs and their patients in France showed significantly lower consumption of antibiotics and antipyretic/anti-inflammatory drugs, yet their symptoms improved at a similar rate. Data was used to compare the effectiveness of antibiotic and antipyretic/anti-inflammatory drugs with additional or alternative use of homeopathy in 518 adults and children with URTIs. The difference between the homeopathy GP group and the conventional medicine GP group may be due to specific attributes of either physicians or patients or interactions between the two.


Grimaldi-Bensouda L, Begaud B, Rossignol M, Avouac B, Lert F, Rouillon F. Management of upper respiratory tract infections by different medical practices, including homeopathy, and consumption of antibiotics in primary care: the EPI3 cohort study in France 2007-2008. PLoS One [Internet]. 2014;9. Available at:


This is an observational study, so it is not designed to say anything about whether homeopathy is effective. It was funded by French homeopathy manufacturing giant, Boiron, but carried out by an independent company that does cohort studies and epidemiological research for drug companies. There is plenty of scope for bias, being a cohort study in which patients decided whether they wanted to see a conventional or homeopath GP, and with much of the data being based on self-report by patients on follow-up telephone interviews. Some of the data came directly from doctors prescriptions, so that seems likely to be reliable.

The more reliable data from doctors show, in France, much less use of antibiotics by the certified homeopaths compared to the conventional doctors, to treat upper respiratory tract infections (URTIs). From the less reliable self-reported patient data, patients who choose GPs who are certified homeopaths rather than conventional GPs, say they are roughly 60% less likely to use antibiotics, and roughly 50% less likely to use anti-pyretic/anti-inflammatories for acute URTIs, but get better (as assessed by self-report one-month later) just as well. This difference could be due to patient or doctor characteristics. However, there is a trend (the study is underpowered because of small numbers) for there to be more follow-on infections (otitis/sinusitis) in the following 12 months in the homeopathic-GP group that used less antibiotics.

However, this finding has nothing to do with whether or not homeopathic remedies (or antibiotics) are effective for treating URTIs. There was no placebo group, the URTIs being assessed were acute, mostly self-limiting, conditions, many patients were not treated with either conventional drugs or homeopathy, some of the homeopathic doctors’ patients were treated with antibiotics, and the 'response to treatment' was assessed one month later when most would be expected to be better with or without treatment. The homeopathic doctors’ patients also had milder symptoms at baseline than the conventional doctors’ patients.

It is well known that doctors over-prescribe antibiotics for acute, self-limiting URTIs (France is among the worst EU countries for this). They could certainly prescribe less antibiotics with little or no detriment to their URTI patients. Homeopathic remedies are probably irrelevant (other than any placebo effects), but it is clear that expectations/attitudes of homeopathic patients and/or their homeopathic doctors do lead to less unnecessary prescribing of antibiotics. But does that constitute a reason to use homeopathic remedies as placebos, rather than just reducing the use of antibiotics?

Martin Whitehead

Paper 5

SoH headline

Patient satisfaction can be higher in primary homeopathic care than in conventional care


Patient satisfaction was significantly higher in the homeopathic care group in an observational study which investigated patient satisfaction and perception of side effects using homeopathy compared with conventional care. This Swiss study, conducted in 2002–2003, used data from two large, cross-sectional studies of over 6,000 primary care patients. There were significant differences in the initial health status of the groups, with a higher percentage of chronic and severe conditions found in the homeopathic group. The authors concluded that homeopathic treatment was perceived as a low risk therapy by patients with two to three times fewer side effects than conventional care.


Marian F, Joost K, Saini KD, von Ammon K, Thurneysen A, Busato A, et al. Patient satisfaction and side effects in primary care: An observational study comparing homeopathy and conventional medicine. BMC Complement Altern Med [Internet]. 2008;8(1):52. Available at: es/10.1186/1472-6882-8-52


This study, based on two questionnaires sent to patients and practitioners, tells us nothing about whether homeopathy works or not. The authors compared two groups of patients, one treated by conventional practitioners (CPs), the other treated by homeopathic practitioners (HPs); practitioners were paid roughly 300 Euros for participating. On the face of it this might seem a straightforward comparison of homeopathic vs conventional treatment, but in reality, the two groups are so hopelessly imbalanced any conclusion is rendered meaningless. For one thing, the HPs were also permitted to use conventional treatments (in the end in only 44% of HP consultations were homeopathic methods used as the sole treatment) yet the patients in these cases were still classified as being in the homeopathic group. Additionally, the HPs were more likely to be female, to practice in city, rather than rural clinics and were more often part-time.

Consultations with HPs were 70% longer on average than those with CPs and the authors concede the HPs would be more highly motivated in favour of homeopathy as would those of their patients who responded to the questionnaires, who were more likely to be younger, female, better educated and more likely to be suffering from chronic disease than their counterparts in the CP patient group. To quote the paper, ‘HP physicians were under pressure to demonstrate effective methods — which was not the case for CP physicians’.

Regarding the SoH assertion, it is misleading to suggest it was homeopathy that resulted in the difference between the two patient groups given that 56% of patients treated by the HPs were treated using non-homeopathic methods and there were a significant number of other differences between the two patient groups and their practitioners, any one of which could have influenced the overly optimistic conclusion. And in the final outcome although the patients who were treated by a HP (though not necessarily given homeopathy) showed greater levels of satisfaction with their practitioner they actually described a lower degree of symptom relief than those in the CP group, something the SoH has failed to mention.

Niall Taylor

Paper 6

SoH headline

Homeopathic treatment compares well with some antidepressants in the treatment of depression in menopause


A recent randomised controlled trial in Mexico concluded that homeopathic treatment compares favourably with that of fluoxetine in the treatment of depression during menopause. The study compared individualised homeopathy with Fluoxetine in the treatment of major depression in 133 peri- and post-menopausal women. After six weeks of treatment, those in the homeopathy group but not the fluoxetine group had improved more than the placebo group on two out of three inventory scales. The menopause symptoms of those in the homeopathy, but not the fluoxetine arms, also improved.


Del Carmen Macías-Cortés E, Llanes-González L, Aguilar-Faisal L, Asbun-Bojalil J. Individualised homeopathic treatment and fluoxetine for moderate to severe depression in peri- and postmenopausal women (HOMDEP-MENOP study): A randomised, double-dummy, double-blind, placebo-controlled trial. PLoS One. 2015;10(3). Available at: ne.0118440


After six weeks of treatment, the depression scores in both the homeopathy and fluoxetine groups improved more than the placebo group on one inventory scale, but neither homeopathy or fluoxetine improved the depression scores on another inventory scale. The scores for menopause symptoms of those in the homeopathy, but not the fluoxetine arms, also improved more than for the placebo group.

This is a well-designed trial — by a long distance the best of the 10 studies selected by SoH. However, it is a small study with only 43 to 46 patients in each group. Overall, previous trials of homeopathy for psychiatric conditions have failed to produce convincingly positive results, so this small trial needs replication to be confident of its results.

Martin Whitehead

Paper 7

SoH headline

Homeopathic treatment could play a role in blood sugar control in patients with Type 2 diabetes


In a Hong Kong study, 27 patients with Type 2 diabetes mellitus given individualised homeopathic treatment plus conventional treatment showed showed [sic] significantly better changes in FPG levels after 12 months than the control group. This is despite the fact that the homeopathic group came to the study with more established and severe symptoms. While the study sample was too small to reach conclusions, given the global health and cost implications of type 2 diabetes, results indicate that follow up studies in this area should be seriously considered.


Lun K, To A, Ying Y, Fok Y, Chun K, Chong M, et al. Individualised homeopathic treatment in addition to conventional treatment in type 2 diabetic patients in Hong Kong e [sic] a retrospective cohort study. Homeopathy [Internet]. 2017;106(2):79-86. Abstract available at:

[Note that the names of the authors given by the SoH are wrong. The correct names are To, K.L.A., Fok, Y.Y.Y., Chong, K.C.M., Lee, Y.C.J. and Yiu, L.S.S.]


This study from Hong Kong is essentially a retrospective chart review where the data of a small number (n=27) of patients with type 2 diabetes who had been given individualised homeopathy in addition to conventional anti-diabetic therapies were compared to the data of two groups of historical controls who both had only received conventional treatments. The authors observed that "homeopathic treatment was associated with better glycaemic control compared with standard conventional treatment alone." However, the study has several serious flaws that render this conclusion unsafe and arguably dangerous. For instance, the homeopathic group received plenty of extra care and attention that is likely to impact the outcomes.

More importantly, the two control groups were not randomised and therefore are likely to differ in all sorts of variables that influenced the outcomes. The baseline characteristics already reveal some potentially important group differences, but there could be many more. For example, it is reasonable to assume that receiving extra attention prompted the homeopathic group to better adhere to their conventional treatments than was the case with controls. Therefore, the observed outcomes are likely to be unrelated to the effects of the homeopathic remedies given, and it could endanger the health of diabetic patients to conclude otherwise.

Edzard Ernst

Paper 8

SoH headline

Homeopathy could be an effective and acceptable treatment for depression


Evidence from a recent RCT suggests that patients with self-reported depression find homeopathy to be an effective and acceptable treatment. The study (a cohort multiple randomised control trial) tested the effectiveness of add-on treatment from homeopaths compared to usual care alone, over a period of 12 months. Results showed a moderate treatment effect size in favour of those treated with homeopathy, which was still present at 12 months.


Viksveen P, Relton C, Nicholl J. Depressed patients treated by homeopaths: a randomised controlled trial using the ’cohort multiple randomised controlled trial’ (cmRCT) design. Trials [Internet]. 2017;18(1):299. Available at:


Homeopathy improved self-reported depression scores, which may indicate a genuine improvement in depression in this unblinded study, or maybe the people filling in the depression scale wanted to please the nice people running the study who had provided them with extra treatment. Effect size is small and may not be clinically significant, and is in addition to the effect (for unknown reasons, not given in the study) of concurrent conventional treatment. If real, there is no evidence the improvement is due to a specific effect of the remedies — it may all be associated with the multiple hour-long homeopathic consult sessions the patients received and/or placebo effects.

The name ‘cohort multiple randomised control trial’ is perhaps misleading to those who don’t know what such a trial consists of. It is randomised, but is not blinded and not placebo controlled; the comparison is conventional treatment + homeopathy vs conventional treatment alone (A+B vs A). This type of trial design will almost always produce a positive result for homeopathic treatment of psychological problems regardless of whether the treatment being tested is actually effective because of the additional attention given to those receiving the homeopathy.

Martin Whitehead

Paper 9

SoH headline

Homeopathy can help to improve the overall health and wellbeing of cancer patients when provided in addition to conventional therapy


Homeopathy may play a role in the treatment of multi-morbid patients, with lower cost implications than conventional treatments. Frei et al treated 50 patients with multiple conditions using polarity analysis - a system based on grading found in Boenninghausen’s Therapeutic Pocket Book. 43 patients stayed for the study period. The cost of homeopathic treatment was less than half the projected equivalent of conventional treatment.


Frei H. Homeopathic treatment of multimorbid patients: A prospective outcome study with polarity analysis. Homeopathy [Internet]. 2015;104(1):57–65. Available at:


The paper by Frei simply describes the natural history of a group of patients following a particular style of homeopathic treatment with the case history of a single patient described in detail. There is no control group, so one can conclude literally nothing about any possible real effect of homeopathic remedies themselves. The changes described are, inevitably, a combination of biased observations hopelessly confounded by the natural history of the various disease processes.

Once again we are faced with the recurrent tendency of homeopathic organisations to use uncontrolled observations as if they have anything at all to tell us about the purported effects of the remedies themselves. It is not that we contend that the reported content of studies such as this is in any sense faked, but that even if accurately documented it tells us nothing useful.

Simon J Baker

Paper 10

SoH headline

Homeopathy may play a role in the treatment of patients with two or more conditions


The results of this pragmatic RCT suggest that the global health status and subjective wellbeing of cancer patients improve significantly when adjunct classical homeopathic treatment is administered in addition to conventional therapy. 410 patients being treated by standard anti-neoplastic therapy were randomised to receive or not receive classical homeopathic therapy in addition.


Frass M, Friehs H, Thallinger C, Sohal NK, Marosi C, Muchitsch I, et al. Influence of adjunctive classical homeopathy on global health status and subjective wellbeing in cancer patients – A pragmatic randomised controlled trial. Complement Ther Med. 2015;23(3):309-17. Available at:


The Frass study is described as a pragmatic randomised controlled study. The most important feature that completely undermines its conclusions is that there is no blinding. Participants were informed of the purpose of the study prior to group allocation so recipients of homeopathy were already primed to perceive a benefit.

The participants allocated to homeopathy had an initial 60-minute session with a homeopath followed by two more 30-minute sessions. Nothing equivalent was done for the comparator group (one cannot really call it a control group in the proper sense of the term). Participants were then asked for their subjective impressions of homeopathic treatment, which risked obtaining answers tuned to please the study organisers rather than genuinely reflecting the experiences of its participants.

The homeopathic group was thus, primed to report a benefit from homeopathic treatment, given three sessions that amounted to extended counselling sessions and then required to comment only on their subjective impressions rather than being tested in any objective way. If one was looking for a study design that maximises bias and creates evidence of what is loosely called a placebo effect, but which is actually an artefact of poor study design, then one could hardly do better.

The authors themselves say, “The study was not designed to show which component of homeopathic treatment was effective.” Indeed not: it appears to have been carefully designed to show an effect from the counselling associated with homeopathy while sidestepping the issue of whether the pills themselves have any effect.

Simon J Baker


Alan Henness, Zeno’s Blog @Zeno001, director of the Nightingale Collaboration, @NightingaleC

Danny Chambers BVSc MSc MRCVS @DannyVet Langford Vets, University of Bristol

Edzard Ernst MD PhD FMedSci FRSB FRCP FRCPEd Emeritus Professor of complementary medicine, Exeter University @EdzardErnst

Gold @Unifex

Martin Whitehead BSc PhD, BVSc CertSAM MRCVS Chipping Norton Veterinary Hospital

Niall Taylor BVM&S MRCVS RationalVetMed

Simon J Baker MA VetMB PhD MRCVS


[1] Note that this references a blog post written by Alan Henness and Sven Rudloff with translations from Swiss-German by Sven Rudloff. However, all original documents are cited, so the veracity of translations (and other information) can be fully verified.

11 December 2017